Views
No views yet
pip install mlx-embeddings1from mlx_embeddings import load, generate
2import mlx.core as mx
3
4model, tokenizer = load("mlx-community/bge-small-en-v1.5-bf16")
5
6# For text embeddings
7output = generate(model, processor, texts=["I like grapes", "I like fruits"])
8embeddings = output.text_embeds # Normalized embeddings
9
10# Compute dot product between normalized embeddings
11similarity_matrix = mx.matmul(embeddings, embeddings.T)
12
13print("Similarity matrix between texts:")
14print(similarity_matrix)
15
16